DTIC AD0424662: DIFFERENTIATION OF AMINO ACIDS BY GAS-LIQUID CHROMATOGRAPHY OF THEIR PYROLYSIS PRODUCTS

UNCLASSIFIED

AD NUMBER

AD424662

NEW LIMITATION CHANGE
TO

Approved for public release^ distribution
unlimited

EROM

Distribution authorized to U.S. Gov't,
agencies and their contractors;
Administrative/Operational Use; Nov 1963.
Other requests shall be referred to Army
Biological Laboratories^ Federick, MD.

AUTHORITY

BDRL, D/A Itr, 27 Sep 1971

THIS PAGE IS UNCLASSIEIED

DEFENSE DOCUMENTATION CENTER

FOR

SCIENTIFIC AND TECHNICAL INFORMATION

CAMERON STATION. ALEXANDRIA. VIRGINIA

UNCLASSIFIED

NOTICE: When government or other drawings, speci¬
fications or other data are used for any purpose
other than in connection with a definitely related
government procurement operation, the U. S.
Government thereby incurs no responsibility, nor any
obligation \rtiatsoever; and the fact that the Govern¬
ment may have formulated, furnished, or in any way
supplied the said drawings, specifications, or other
data is not to be regarded by implication or other¬
wise as in any manner licensing the holder or any
other person or corporation, or conveying any rlfi^tits
or permlaston to manufacture, use or sell any
patented invention that may in any way be related
thereto.

•• •> •••V* ******** *■ • *

• 1 • 6 • • ■ * • d

:u::

tttiir.xv^rsf

Jiiii J

• •••••••••i****** ^ -CP*

' ■ I ifV> lit* 1 1 I « 1 1

> v**^***?********

.•••fc***********

• •••••a***********

.«»«»^i|ip****«* ^

ilSSiSSjSiSSK******

^NV*^^***^******

. «•••••••»•*••««*•«

•••*»••••••«••••«•

.^;::::::s:n:::s:9

>«••••*••••••••••••

<*••••>*•*»•••«••••«
. •••f *««*«i***«*«*4

•'•*«»*«*****«**4*«*

•«••*••••••••»■••«•

.•••••*••••«••«*•*•*

' !*«*«**«S*S85*»****

. is#sss*eetf?»»^'^?***

- (••••*•••••••••••••

• •••»••••••••*••*••*

.«••«»*••*••«•*••••*
>*»e»**** »•**•••*•**
w »••••• ****4********
> •••*•*•••••*•*««**•
I ••i****************
. •••»•••«•*»•«•••••«
• r******************
*•••*«••••••«••••*•*

■ •••«««**«**«M*«*<

*•••«••**«••••••*••«

• •••«'»W«**«********

Hi

_ _ _

t**«**»#******f ******
* *••»••••••««*•**«•*«

‘»*********^4*tit!tllJ

»•***•••■♦*••*•*••••••

-•••••**•••***•******

.«•••••••«•*««•*«*•••

.:**8«*«*«**********«

^V***********-********

■••••••**•*•••*••*•••

a •*•••*# •****2;******

^*«****«

*•••••

_ •«••*•

_ _

•••♦••••••I**********

(***••**>*****''

••••««•••••****■<>****

,♦**•••••••*•••*♦*;•♦

•«««••*•»•***•*•*•**

••••••••••**••«*••«

•«**••••«*•*•**•*•

••♦•••»*«••••«***

•••*•*••«««**•*•

•n************

^••«**«»*****

•••**•*••••*

««***•«»*•*

DIFFERENTIATION OF AMINO ACIDS
BY GAS-LIQUID CHROMATOGRAPHY
OF THEIR PYROLYSIS PRODUCTS

NOVEMBER 1963

UNITED STATES ARMY
BIOLCX3ICAL LABORATORIES
FORT DETRICK

U.S, ARfft: BIOLOGICAL LABORATORIES
Fort Deti'ick, Frederick, Marykind

TECHNICAL MANUSCRIPT 87

DIFFERENTIATION OF AMINO ACIDS BY GAS-LIQUID
CHROMATOGRAPHY OF THEIR PYROLYSIS PRODUCTS

t

Leonard N. Winter
Phillip W. Albro

Physical Defense Division
DIRECTOR OF MEDICAL RESEARCH

Project 1C622401A071

November 1963

1

Pcitions of the work reported here were
performed under Projects 4B11-05-014 and
1C022301A071^ "Research on BW Rapid Warning
System," Task -01 "Physical Principles of
Detection." The expenditure orders .were
2017 and 5330. This material was originally
submitted as manuscript 5240.

DDC AVAILABILITY NOTICE

Qualified requestors may obtain copies of this
document from DDC .

Foreign announcement and dissemination pf this
document by DDG is limited.

The Information in this document has not been
cleared for release to the public.

ABSTRACT

Conditions are described for the low- temperature pyrolysis
of amino acids and gas-liquid chromatography of the amines
produced. Pyrolysis is accomplished at 300°C, and the amines
are stabilized at 110“C prior to chromatography on Quadrol,
Aliphatic amines C^ to C5 can be observed under these conditions.
Each amino acid gives a unique amine profile^ and proteins give
reproducible amine profiles related to their amino acid content.

CONTENTS

Ahstra.c-l:

3

I. IRTRODUCIION

II. EXPERIMENTAL WORK. .

A, Apparatus . .

B. Column Preparation .

1. Column Packing .

2. Conditions for Pyrolysis . .

3. Conditions for Chromatography

G. Procedure . .

5

5

5

7

7

7

8

III. RESULTS AND DISCUSSION

11

Literature Cited

19

FIGURES

1. Pyrolysis Chamber . . 6

2. F4ffect of Filament Power on Evolution of Amines from Egg Albumin . 9

3. Chromatograms of Egg and Bovine Serum Albumin Pyrolysates .... 12

4. Chromatograms of Phenylalanine-Methionine Mixture Pyrolysate ... 12

TABLES

I. Identification of Peaks in Tables . 10

II. Amine Profiles' for Proteins and Peptides . 10

III. Comparison of Computed and Measured Peak Heights for Egg Albumin

Pyrolysate. . . . . .... 13

. 14

IV. Amine Profiles of Amino Acid Pyrolysates . .
V. Effect of Py.; ulyzing a Mixture . .

16

5

I. INTRODUCTION

Amino acids of protein h^drolyzates, usually identified by paper chroma¬
tography or electrophoresis, may be quantitatively analyzed by ion exchange
techniques,^’® gas chromatography of their volatile derivatives, or by
gas chromatography of their catalytic oxidation products/ This report
describes a technique for the study of amino acids by gas chromatography of
their low-temperature pyrolysis products. Evidence indicates that this
technique may be applied to the identification of specific proteins and
other nitrogenous material.

II. EXPERIMENTAL WORK

A. APPA'IATUS

The glass fiber filter paper used in these experiments was obtained
from Mine Safety Appliances Company, Pittsburgh, Pennsylvania. The amino
acids and peptides were purchased from the California Corporation for
Biochemical Research, Los Angeles 34, California, the Mann Research
Laboratories, Inc., New York 6, New York, and the Nutritional Biochemicals
Corporation, Cleveland, Ohio. The proteins were obtained from Nutritional
Biochemicals Corporation, Merck and Company, Rahway, New Jersey, and from
Difco Laboratories, Detroit, Michigan. The gas chromatograph was an F &

M Model 500 with a tungsten filament detector. The pyrolysis chamber was
designed by the authors for simple use with the Model GV-11 gas sampling
valve of this instrument, and, is diagrammed in Figure 1.

B. COLUMN PREPARATION

Unpublished work at the U.S. Army Biorogical Laboratories* indicated
that the aiain nitrogenous pyrolytic products of protein at a temperature
of approximately 300°C are ammonia and aliphatic amines. Many problems
are encountered in the gas chromatography of Amines,® including "tailing"
due to adsorption on the solid support and to the extensive solubility of
amines in the water produced during pyrolysis. Water tends to condense
in the connective tubing of the gas sampling valve, and may prevent some
of the amine material from ever reaching the column. Moreover, water
usually is eluted from , the column in the region of interest, thereby
obscuring some of the, amine peaks on the chromatogram.

* Randall, G. U.S. Army Biological Laboratories, personal communication.

6

b

-e-i

Figure 1, Pyrolysis Chamber.

a = 1" X 1" X 1%" stainless steel chamber
b = 1/8" O.D. stainless steel tubing
c = clamp for plug
d = ammeter

e = 18"gauge copper wire leads

f = Teflon plug

g = tungsten filament

h = sample

1 => carrier gas In

j = carrier gas out

Condensation in the connective tubing was minimized by treating the
inside surface with hexamethyldisllazane^ and by maintaining the tern*
perature of the tubing above 100*C with a heating tape. Several methods
for deactivating the solid support and retarding the elution of water
have been previously suggested.®’® These were found unsatisfactory for
the present purposes.

Columns of 0.75 per cent silicone gum rubber and of 20 per cent
triethanolamine on Anakrom ABS were found completely unsatisfactory;
amines tailed badly and water interfered. Thirty per cent of a mixture
of 15 per cent Nujol and 85 per cent hende.canol on Gas Chrom p®~^
was satisfactory for ammonia and methyl amines, but the useful life of
the column (about two weeks) was too short for routine use.

It was found that columns of five per cent DC silicone oil 710 or
five per cent sllilcone oil 200 on methanolic K0H"WaBhed Chromosorb W,
programmed from 100”C to 200°C at 7.9 degrees per minute, were excellent
for the separation of to Cg aliphatic amines and pyridine homologues.
However, few amines above C4 were produced in the amino acid pyrolysates.

7

I . Ooiinuu .Pr.i(;king

'I'he column packing found most satififactory fot separation of air,
ammonia, and Co aliphatic amines with no interference from water was
prepared as follows;

Five grams of reagent-grade KOH were dissolved in enough distilled
water to slurry 9!) grams of sllanized Chromosorb P (120/140 mesh).
After drying at 110"C, the ICOH-coated Chromosorb was slurried with
17.7 grams of Quadrol [tetrakls (2-hydroxypropyl) ethylenedlamlnej
ip a five per cent solution of chloroform in light petroleum ether.
Thi.9 provided 0,15 gram of Quadrol per gram of packing. After the
solvent evaporated, the packing was cured at 105°C overnight and
packed into a copper or stainless steel tube six feat long and four
raillimetera inside diameter. The ends of the tube were secured with
small plugs of hexamethyldisilazane-treated Pyrex glass wool to avoid
amine adsorption at this often-neglected point. If copper tubing is
used, its inside surface should also be silanized. No decomposition
of amines was observed when this was done. This column was operated
at 70°C, but the temperature was raised to 100“C between analyses to
remove adsorbed water,

2. Condltibns for Pyrolysis

Pyrolyzer - see Figure 1
Volte - 6 a.c.

Amperes - 6

Pyrolysis Time - 3 minutes

Filament Temperature - 950°C based on color

Chamber Atmosphere - helium

Injection (chamber flush) - 10 seconds onto column

3. Conditions for Gas Chromatography

Column - 157i. Quadrol over 5% KOH on sllanized Chromosorb P,

120/140 mesh
Length - 6 ft.

Inside Diameter - 4 mm.

Carrier Gas - Helium filtered through Linde Molecular Sieve 13X
Column Temperature -
Detector Temperature - 125°C
Detector Current - 150 mamp.

C. I'ROCEDDRE

A 30“mllligram sample of amino acid or protein was wrapped in glass
fiber filter paper and inserted into the filament of the pyrolyzer as shown
in Figure. Ij rather than being coated on the filament in the. usual manner.''^'""
The wrapping Insured that the sample would not be driven off the filament
before, pyrolysis was complete. The chamber was tested for leaks and purged
with carrier gas while the column was held at 105"C to prevent the adsorption
of moisture from the air in the chamber. The resulting helium atmosphere
in the chamber prevented the oxidation of the amines to nitrogen oxides^
which would interfere with the analysis.

After the chamber was purged and the carrier gas diverted past it, the
column temperature was readjusted to 70“C. This temperature gave the best
resolution of the amines studied. The sample was then heated to SOO^C as
determined by a thermocouple probe Inserted into the filter paper, and
current was maintained through the filament for three minutes to complete the
evolution of degradation products and to allow for equilibration of the gases
at the temperature of the chamber atmosphere (110°C). This equilibration
allowed Interaction of the vapors, producing a more complex amine mixture
than would have been stable at the filament temperature. It was also found
necessary for reproducible profiles. Slight variations In filament tempera¬
ture were found to significantly affect the quantitative yield of amines,
but the qualitative yield was constant over a range of about 100 degrees
(Figure 2). '

The carrier gas stream was channeled through the pyrolysis chamber,
flushing its contents onto the column for ten seconds. The gas stream was
then re-routed past the chamber. Although this did not completely empty
the chamber of pyrolysate vapors, it gave the maximum sample size that would
not overload the column.

The amines were Identified by comparison of their relative retention
volumns (tripropylamine = 1,00) with those of known standards analyzed under
the same chromatographic conditions (Table I). The relative retention
volumns obtained from the pyrolysate amines were corrected for the delay in
moving from the pyrolysis chamber to the column. The effluent from the column
side of the detector was passed into methyl red or Nessler's reagent to con¬
firm the presence of amines.

To determine if this technique could be applied to proteins and protein
mixtures, samples of crystalline egg albumin, bovine serum albumin,
glutathione, hlstldylhistidlne and lyophllized cellsof Sarclna lutea were
pyrolyzed under conditions identical to those used on individual amino acids.
Results are shown in Table. II,

1,0

TABLK T. I.DKNTTFICATION OF PEAKS IN TABLES

Re la t Ive

Peak

Compound

Retention

Volume^^

A

Ammonia

0.42

B

Methyl and Dimethyl Amines

0.60

C

Ethylamlne

0.82

D

unidentified

0.92

E

Tripropylamine and/or Benzene

1.00

F

D1 propylamine

1.25

G

Trlbutylamlne

1.46

H

Butylamine

1.96

I

Dlbutylamlne

3 , 66

J

unidentified

3.16

K

unidentified

2.08

L

Amy famine

2.37

a, Tripropylamine set equal to 1.00.

TABLE

11. AMINE

; PROFILES FOR

PROTEINS AND

PEPTIDES

Average Peak Height, inches

Egg

Bovine

Hist idyl-

Glycyl-

Peak

Albumin

Albumin

Hemoglobin

histidine

glycine

Glutathion

A

16.3

40.0

27.2

40.0

45.6

12.8

B

2.0

1,8

1.0

-

-

1.2

C

0,5

4.2

-

-

-

-

D

4.9

~

3.5

-

-

2.5

E

4.2

-

1.7

1.8

7.2

6.1

F

-

4.4

-

-

-

-

G

0.2

0.1

-

0.3

-

-

H

I

j

0.1

-

0.2

0.1

-

-

K

L

-

-

-

~

-

-

1.1

All otl'.o.uipl: V\;as made Lo relate the amine profile of S. Itlliei): io tl't’ aniino
acids actually present. Ono gram of lyophilized S. lutea cells was hydiroly?;''.
in 6N HCl under reflux for three hours. The hydrolyzate was aucilyzed by
paper chromatography against known .standards. N~butanol/glacial acetic acid/
water, 2:1:1, and Isopropanol/con . HCl/wate.r, 65.0 : 16.6 : 18,4 were
used as .solvent systems according to the method of Fink, Klein and Fink.
Serine, cystine, leucine, glutamic acid, phenylalanine, alanine, and tyrosine
'were found present. Cystine, serine, and leucine appeared to be the major
amino acids released in this brief hydrolysis.

Samples of crystalline egg albumin and bovine serum albumin were pyro-
lyzed as described above (Figure 3). An attempt was made to relate the
amine profiles obtained from these proteins to the. amino acids they contain.
The amount of each amine obtained from each amino acid (when pyrolyzed
separately), was multiplied by the per cent of that amino acid in the par¬
ticular protein being studied. ® If the amount of each amine in the protein
pyrolysate was directly related to the sum of the contributions to that
amine made by each amino acid releasing it, summing the (amount) x (%)
values obtained above should give a value equal or proportional to the
amount of that amine in the protein pyrolyzate. The results are summarized
in Table III.

To further test the possibility that each amino acid when pyrolyzed in
a mixture yields the same amines in the same proportions that it yields when
pyrolyzed separately, mixtures of phenylalanine-valine 1:1 (w/w) and phenyl¬
alanine - methionine 1:1 (w/w) were pyrolyzed (Figure 4).

III. RESULTS AND DISCUSSION

Under the conditions described, each amino acid or protein gave a unique
and reproducible amine profile, as set forth in Tables II and IV. No amines
were observed having carbon chains longer than those present in the parent
amino acid. However, the presence of di- and tri- amines Indicates that
mere degradation of the amino acid does not account for all of the amines
observed. Interaction and recombination probably take place in the cooler
portions of the chamber.

Several peaks appeared that did not correspond in relative retention
volume to any symmetrical amine standard. Some of the unidentified peaks,
for example peak J (Table IV) from leucine, elute from the column at points
expected for Iso-amines. Since one amine usually present in each elution
profile .Is equivalent to the parent amino acid less the carboxly group.

n

TABLE III. COMPARISON OF COMPUTED AND MEASURED
PEAK HEIGHTS FOR EGG ALBUMIN PYROLYSAXE

Peak

Samp le 1

Sample 2

Sample 3

Average

Computed

A

18.4

16.0

14,6

16.3

16.2

B

2.2

1.8

, a/
NR-

2.0

2.0

C

0.7

0.5

0.4

0.5

0.4

D

5,1

4.8

4.9

4.9

4.0

E

4.0

4.4

4.3

4.2

4.2

F

NOT

RESOLVED

FROM E

-

0.8

G

0.2

0,3

0,0

0.2

0. i

H

0.1

0.1

0.1

0.1

0.1

I

0

0

0

0.0

0.0

1 J

1

0

0

0

0.0

0.0

1

! K

p

0

0

0.0

0.0

; ■ , L

/

0

0

0

0.0

0.0

1 !.

a . NR =»

not resolved to

automatic

attenuation

on recorder.

14

TABLE IV. AMINE PROFILES OF AMINO ACID PYROLYSATES

Amino Acid

Peak Height,

inches

iS:/

A

B

0

D

E

F

G

H

I

J

K

L

Phenylalanine

18.8

1.9

7.0

-y

5.3

1.6

0.2

-

-

0.5

-

-

Glyc Ine

53.8

-

-

-

6.1

-

-

-

-

-

-

-

Serine

30.0

-

~

1.1

5.6

0.7

0.2

-

"

-

-

-

Asparagine

76.0

1.3

-

-

1.6

-

-

'

-

-

-

-

Nor leucine

22.4

1.3

1.0

0.4

0.9

0.4

-

-

-

0.6

-

-

Cystine

12.0

-

-

1.4

-

4,5

-

-

-

-

-

-

Threonine

24.0

5.2

1.2

1.1

0.9

0.2~^

0,3

-

-

-

-

-

Aspartic Acid

1.3

-

-

-

-

0.8

-

-

-

-

-

-

Glutamic Acid

-

1.1

“

3,0

3.1

-

-

-

-

-

-

-

Alanine

24.0

-

-

27.2

14.0

-

0.4

-

-

-

-

-

Valine

16.0

18.0

-

2.6

6.8

-

-

-

-

-

0.5

-

Isoleucine

8.0

0.1

-

0.4

3.7

0,5

0.6

-

-

-

0.7

-

Leucine

16.0

1.2

0.6

1.2

4.3

1.0

-

-

-

1.3

-

-

Pro line

3.0

1.0

-

■ -

0.5

-

-

-

-

-

-

Tryptophan

-

1.9

-

2.1

-

-

-

-

-

-

-

-

Lysine

1.6

1.6

1.5

6.1

1.2

0.2

0.3

-

>

-

-

-

Methionine

38.4

1.5^

^2.5

6.1

11.6

-

“

-

-

-

-

-

Tyrosine

30.8

1.6

-

0.9

-

0.1

-

-

-

0.1

0.1

-

a. From 30"milllgram amino acid samples, average of three determinations.

b. Peak absent,

c. Occasionally much higher; very temperature-sensitive.

15

tlip. structure of leuciue suggests that: isobutyl amine slioiild be prcseni iu
its pyrolysate, H iwever, positive identi tication of thi.s peak ran nul be
made, in the absence of known iso-amine standards^ especially since, peak J
also results from the pyrolysis of tyrosine, which contains no iso groups.

The amines found In the pyrolysate of _S. lutea included all the. amines
observed in the profiles of the amino acids known to be present. No other
peaks appeared.

It is apparent from an examination of Table III that the. amine profile,
obtained from egg albumin is related to the individual amine profiles of
the component amino acids. In the case of some of the amines, the relation¬
ship appears to involve simple addition of the amounts of those amines that
would be produced from each amino acid if pyrolyzad separately. In other
cases, for example peak D, the relationship does not appear to involve simple
addition. One fact involved in this observation may be that benzene, a
possible pyrolytic product of phenylalanine and tyrosine, is also eluted at
peak D.

Several samples of egg albumin, each from a different source, all gave
qualitatively Identical amine profiles. However, Figure 3 shows that even
proteins of the same general class, in this case egg and serum albumins,
exhibit significant qualitative differences when analyzed by this technique.
It is interesting to note that peak F is seen to be much more conspicvious
in the serum albumin profile than in that of egg albumin. When pyrolyzed
alone, cystine yields large quantities of amine eluting at peak F, and
cystine makes up four times as much of the amino acid content of serum
albumin as it does of egg albumin.

When a mixture of phenylalanine and methionine was pyrolyzed, the
resultant amine profile was qualitatively similar to a combination of the
profiles obtained by pyrolyzing phenylalanine and methionine separately.
However, the amounts of each amine in the pyrolyzed mixture were, not equal
to the sums of the amounts that would have been obtained by separately
pyrolyzing the components (Table V), Moreover, when the phenylalanine-valtne
mixture was pyrolyzed, the resulting amine profile contained amines not
found in the pyrolysates of either Individual amino acid.

A comparison of the results of pyrolyzing glycine (Table IV) with those
of pyrolyzing glycylglycine (Table II) shows that the two amine profiles are
qualitatively similar, and moreover that the ratio of total amine from
glycylglycine to total amine from glycine is directly proportional to the
ratio of their nitrogen contents.

TABL,E V. EFFfcXiT OF PYROI.YZING A MIXTURE

Teak

P*

%(P+M)

Peak Height, Mixture^

A

18.8

38.4

27.6

13.1

B

1.9

27.2

14.5

11.2

C

7.0

2.5

4.8

4.5

D

-

5.2

2.6

2.2

E

5.3

10.6

8.0

9.0

F

10.1

5.1

5.3

G

0.2

~

0.1

0.15

H

-

-

-

■ -

JL

J

0.5

-

0.25

0.2

K

-

-

L

-

-

-

- .

* Peak heights for 30 milligrams phenylalanine.

+ Peak heights for 30 milligrams methionine.

X Actual peak heights for mixture of 15 milligrams phenylalanine and
15 milligrams methionine.

n

Every suhslance containing amino acid so far Lcsf.'od by raeth(}d ba;'.

given a unique amine, profile. Aithough some, information can be clednced
from tliis profile as to the structure and composition of the cx,ibsl;ance
pyrolyzed, this technique is presently most useful as a means of identifying
particular nltrogenoiis compounds. With further theoretical study and
refinemontc for quantitative work, this technique may be found useful in
the future study of such Incompletely characterized substances as enzymes,
nucleic acids, and antibodies.

19

LT'FERATURl!: CITED

1. Block., R.J.; Dui'cum, E.; and Zwetg, G. "Paper chroma ir.ography and
paper electrophoresis," 2nd ed,. New York, Academic Press, 1958.

2. Spademan, D.H.; Stein, W.H. ; and Moore, S. "Automatic recording
apparatus for use in the chromatography of amino acids," Anal.

Chem. 30:1191-1206, 1958.

3. Sainuelson, 0. "Ion exchangers in analytical chemistry," New York,
John Wiley & Sons, Inc., 1953.

4. Johnson, D.S,; Scott, S.J.; and Meister, A. "Gas-liquid chrom¬
atography of amino acid derivatives," Anal, Chem. 33:669-673,

1961.

5. Marco, G.; Zomzely, C.; and Emery, E. "Gas chromatography of the
n-butyl-n-trif luoroacetyl derivatives of amino acids," Anal. Chem.
34:1414-1417, 1962.

6. Landowne, R.A., and Llpsky, S.R. "Ultrasensitive analysis of amino
acids via gas chromatography and electron capture spectrometry,"
Fed, Proc. 22:235, 1960.

7. Zlatkis, A.; 0r6, J.F.; and Kimball, A.P. "Direct amino acid
analysis by gas-liquid chromatography," Anal. Chem. 32:162-169,
1960.

8. Sze, Y.L.; Borke, M.L.; and Ottenstein, D.M. "Separation of lower
aliphatic amines by gas chromatography," Anal. Chem. 35:240-242,
1963.

9. Smith, E.D., and Radford, R.D. "Modifications of gas chromatography
substrates for the separation of aliphatic diamines,"’ Anal. Chem.
33:1160-1162, 1961.

10. James, A.T.; Martin, A.J.P.; and Smith, G.H. "Separation and
micro-estimation of ammonia and methyl amines," Blochem. J.
52:238-242, 1952.

11. James, A.T. "Gas-liquid partition chromatography: The separation of
volatile aliphatic amines and of the homologues of pyridine,"
Blochem, J. 52:242-247, 1952.

20

12. Hewitt, B.C., and Whithamj B.T. "The identification of suhstances of
low vola.tllity by pyrolysis/gas chromatography/' Analyst 86:643-652,
1961.

13. Janak, J., "Indentif ication of the structure of tionvolatile organic
substances by gas chromatography of pyrolytic products^" Nature 185:
684--686, 1960.

14. Finkj K.; Kline^ R.; and Pink, R. "Paper chromatography of several
classes of compounds: Correlated Rj. values in a variety of solvent
systems," Anal. Ghem. 35:389-398, 1963.

15. Hawk, P.B.; Oser, B.L.; and Summerson, W.H. "Practical physiological
chemistry," New York, The Blaklston Contpany, 1947- p. 109.